Fractures due to radiation-induced bone loss are a recognized adverse effect of radiation therapy (RT). Treatment of such fractures is challenging, with delayed union and nonunion rates in excess of 50%. Quantitative biomarkers of radiation-induced bone loss are needed to support the design of RT protocols minimizing bone loss and to guide the application of prophylactic interventions. X-ray CT is sensitive to mineralized tissue and thus well positioned to detect bone loss. CT is established as a biomarker of bone composition, estimated via volumetric bone mineral density (vBMD). However, comprehensive evaluation of bone health requires consideration of bone microarchitecture in addition to mineralization. Rapid changes in metrics of bone architecture in response to radiation have been found in pre-clinical micro-CT. Translation of quantitative indices of bone microarchitecture from micro-CT to clinical imaging is challenged by the comparatively coarse spatial resolution of multi-detector CT (MDCT) and flat-panel detector cone-beam CT (CBCT). In the application considered here, the compromised absolute quantitative accuracy in the architectural metrics can be tolerated as long as the biomarker accurately captures relative changes in morphology. A recently developed high-resolution extremities CBCT system was shown to yield better correlation with micro-CT than MDCT across a range of metrics of bone architecture. Accurate measurements of vBMD were also demonstrated, owing to advanced artifact correction. These desirable properties, combined with modest imaging dose, make the extremities CBCT system well suited to test the hypothesis that imaging biomarkers of bone microarchitecture can detect and quantify radiation-induced bone loss in vivo in patients undergoing radiotherapy. The project will involve the following aims: 1) Conduct a clinical pilot study of quantitative imaging in monitoring of bone quality during radiotherapy on twenty patients receiving RT of extremity sarcoma. Both the treated and contralateral (control) limb will be imaged using the extremity CBCT at 4 time points: before RT (baseline), mid-way through RT fractionation, at conclusion of RT, and 3 months after RT. 2) Quantitatively assess radiation-induced bone loss using image-based biomarkers. Changes in vBMD and bone architecture metrics in the irradiated limb will be assessed relative to the control limb. The effects of radiation on trabecular bone will be also quantified using texture analysis. Successful completion of the Aims will demonstrate image-based biomarkers of radiation-induced bone loss in human subjects. The methodology will be developed using extremity CBCT, and the results will inform extension to MDCT and on- board CBCT in image-guided RT. The biomarkers established here will facilitate investigation of the clinical significance of RT-induced changes in bone morphology and support the development of means by which post-RT fractures could be better managed or obviated by optimized treatment planning and adjuvant therapies.